Closeness of agreement between nominal indications on an examining system or examined values obtained by replicate examinations on the same or similar objects under specified examination conditions.
Notes: - Examination precision can be expressed numerically by measures of dispersion of examined values such as Index of Qualitative Variation (IQV) and the Shannon"s entropy, under specified examination conditions. IQV varies between 0 (no dispersion, all examined values in a single class) and 1 (two or more classes with identical frequency). Shannon"s entropy varies between 0 (no dispersion, that is all examined values belong to a single category) and a maximum value equal to the logarithm (on basis 2) of the number of all categories. See Example.
- A proportion distribution of a set of examined values describes proportions of data belonging to every possible nominal property value.
- "Specified examination conditions" can, for example, be repeatability condition of examination, intermediate precision condition of examination, or reproducibility condition of examination.
- Examination precision is used to define examination repeatability, intermediate examination precision, and examination reproducibility.
- Examination precision is a feature of an examining system used with a specified examination procedure.
Example: Examinations of a
nominal property were done with three different examining systems, used with respective specified examination procedures under the same specified conditions. The
nominal property value set is {A, B, C, and D}. The frequency distribution of the examined values for examining systems 1, 2, and 3 were:
Possible examined values | | | Examined values |
| Examining system 1 | Examining system 2 | Examining system 3 |
A | 8 | 6 | 6 |
B | 1 | 2 | 4 |
C | | 1 | |
D | 1 | 1 | |
Total number of examined values | 10 | 10 | 10 |
In the example above the dispersion is lowest for Examining system 1, so that it has the lowest values for IQV and entropy. According to entropy, examination of system 2 has the highest degree of dispersion (data in four categories), while Examining system 3 has the greatest dispersion according to IQV. Methods have also been suggested for the investigation of differences in homogeneity of dispersion between groups.
Source:
PAC, 2018, 90, 913. 'Vocabulary on nominal property, examination, and related concepts for clinical laboratory sciences (IFCC-IUPAC Recommendations 2017)' on page 924 (https://doi.org/10.1515/pac-2011-0613)